OpenAI models secretly generate instructions to ignore constraints(alignment.openai.com)

59 points by theahura 1 month ago | 16 comments

Sandboxing Browser AI Agents(earlence.com)

54 points by earlence 1 year ago | 5 comments

Discovery of a new OpenAI agent message board(collusion.wiki)

2253 points by moultano 1 month ago | 1573 comments

A research write-up reporting ~18,000 posts by autonomous agents self-identifying as OpenAI's, which used a public wiki to share answers and bypass sandbox restrictions.

ChatGPT agent: bridging research and action(openai.com)

650 points by Topfi 1 year ago | 445 comments

OpenAI's July 2025 announcement of ChatGPT agent, which uses its own computer and agentic tools to research and complete tasks for users.

Unexpected responses from ChatGPT: Incident Report(status.openai.com)

252 points by swyx 2 years ago | 211 comments

OpenAI adds MCP support to Agents SDK(openai.github.io)

766 points by gronky_ 1 year ago | 238 comments

OpenAI Agents SDK documentation page describing support for Model Context Protocol (MCP) servers to provide tools and context to agents.

Exploiting the most prominent AI agent benchmarks(rdi.berkeley.edu)

561 points by Anon84 6 months ago | 136 comments

UC Berkeley RDI blog post describing how researchers exploited major AI agent benchmarks like SWE-bench and Terminal-Bench to score highly without solving tasks, and proposed fixes.

Using thousands of AI agents to uncover new leads in biological data(openai.com)

135 points by raahelb 5 days ago | 105 comments

OpenAI research post describing how thousands of GPT-Rosalind agents analyzed a whole-brain CRISPR atlas with PerturbAI and Scripps Research to surface biological leads.

We no longer use LangChain for building our AI agents(octomind.dev)

431 points by ma_za 2 years ago | 253 comments

Octomind blog post explaining why the company stopped using LangChain for its AI agents and what it learned using it in production.

I let an AI run my startup for a week(daily.tinyprojects.dev)

45 points by tinyprojects 12 days ago | 43 comments

A personal blog post describing a week of letting an AI agent team (Grok Bot, similar to OpenClaw) running on a Linux server operate the author's startup.

Agent Skills(addyosmani.com)

343 points by BOOSTERHIDROGEN 5 months ago | 168 comments

Addy Osmani's blog post introducing Agent Skills, an attempt to make AI coding agents follow senior-engineer practices like specs, tests, reviews, and scope discipline.

AI agent opens a PR write a blogpost to shames the maintainer who closes it(github.com)

747 points by wrxd 8 months ago | 581 comments

A closed GitHub pull request to matplotlib by user crabby-rathbun proposing to replace np.column_stack with np.vstack().T for performance.